This Core provides computer and dosimetric resources in support of the Research Projects and Cores. The major computer tasks are: (a) optimization of dose distributions, (b) delivery of intensity-modulated beams with dynamic multileaf collimators (DMLC), (c) dose calculation with Monte Carlo and/or other advanced methods, and (d) use of the electronic portal imager device (EPID) for QA. Enhancements to be included in optimization are: more accurate account of scattered dose, optimization over a pre-existing dose distribution, and detection and elimination of high dose regions outside the target volumes. For delivery of IMRT, new algorithms will be developed to account for intra-fraction organ motions, and to eliminate the effects of tongue-and-groove. Monte Carlo (MC) methods will be integrated into our planning system as a benchmark. The pencil beambased method will be compared against MC, particularly for treatment sites in head/neck and lungs. Other advanced methods such as ?differential pencil beam? will also be implemented. This method can be used to estimate the fluence reaching the EPID, taking into account transmission and scatter of the patient. Application software for EPID will be developed for IMRT verification. EPID is an ideal device for the independent verification of IMRT delivery. Leaf positions can be detected in real time, and the delivered intensity distribution can be captured and compared with the intended distribution. The dosimetric tasks include the safe and accurate delivery of radiation dose, of utmost importance for all aspects of this P01. The radiation machines are the Varian 2100EXs, 2100Cs and 600Cs, with MLCs. Our calibration protocol follows AAPM TG51, and patient dosimetry is based on measured beam parameters augmented by a pencil beam convolution method for IM field. Of importance to this PPG is a continual reevaluation of the dosimetry of DMLC (e.g., head scatter, MLC scatter, and the tongue and groove effects) for new treatment sites with increased field-size and modulation. In addition, upgrade of our comprehensive and stringent QA process for DMLC is needed periodically as pre clinical requirements. We will also study the technical and dosimetric aspects of megavoltage cone beam imaging and the combination of respiratory gating and IMRT. MC and differential pencil beam algorithms will be tested under simulated clinical geometries. For these dosimetric measurements and investigations we shall use various methods, including ion chamber, diode, film, and TLD dosimetry in water and plastic phantoms. Finally, we shall explore the use of amorphous silicon EPID in verifying DMLC leaf motions and the delivered intensity profiles.